Objective To develop a risk prediction model for diabetic foot in patients with type 2 diabetes mellitus （T2DM）, and to provide a reference for medical staff to identify high-risk patients and formulate targeted intervention programs. Methods Clinical information of patients with T2DM hospitalized from March 2018 to March 2020 were retrospectively collected for model development.The risk factors of diabetic foot were identified using LASSO regression and multivariate logistic regression, then a predictive nomogram model was built based on these risk factors.The C-statistic was calculated to evaluate the discrimination of the mo-del, and the calibration plots and Brier scores were utilized to assess the calibration of the model.Internal validation was performed using the bootstrap method, and external validation was performed in another group of patients with T2DM hospitalized between April and December 2020. Results A total of 1,527 patients were included in the development data set, and 319 （20.89%） had diabetic foot ulcers.Nine predictors were eventually determined to be associated with diabetic foot:education level, history of prior ulcers or amputation, retinopathy, injection of insulin, vibration perception threshold, dorsalis pedis artery pulse, foot skin anomalies, fungal foot infections, and foot deformities.The C-statistic of the model was 0.966 without correction and 0. 963 after bootstrap correction.The Brier score was 0.052 and bootstrapping corrected Brier score was 0.054.The calibration plot showed good agreement of the nomogram between predicted and observed outcomes.A total of 451 patients were included in external validation data set, and 71（15.74%） suffered from diabetic foot disease.The C-statistic was 0.928, the Brier score was 0.051, and the calibration plot also showed good agreement between observed and predicted probabilities. Conclusion The nomogram prediction model has good discrimination and calibration, which can be used for screening of patients at high risk for diabetic foot.